Grok Word Limit by Model (2026)
xAI's Grok grew from a 128k context ceiling to 1 million tokens in about a year. Here's what fits in each current model, what it costs, and where the practical limits sit.
Quick Answer
Grok 4.7, the current flagship (September 21, 2026), accepts 500,000 tokens (~370,000 words) at $2 per million input and $6 per million output. Grok 4.6 matches those limits and pricing and remains available. Grok 4.3 accepts 1,000,000 tokens (~750,000 words) at $1.25 per million input. The older Grok 4, Grok 4 Fast, Grok 4.1 and Grok 4.1 Fast (up to 2M tokens) were retired by xAI on May 15, 2026 and now redirect to Grok 4.3.
Grok context windows by model
| Model | Input tokens | Input words | Released |
|---|---|---|---|
| Grok 4.7 | 500,000 | ~370,000 | Sep 2026 |
| Grok 4.6 | 500,000 | ~370,000 | Aug 2026 |
| Grok 4.5 | 500,000 | ~370,000 | Jul 2026 |
| Grok 4.3 | 1,000,000 | ~750,000 | Apr 2026 |
Grok 4, Grok 4 Fast, Grok 4.1, Grok 4.1 Fast and Grok 3 were retired on May 15, 2026 and are no longer listed. Specs from xAI official documentation and OpenRouter model cards as of October 2026. Verify current specs with xAI before relying on these for production.
How Grok caught up on context length
Grok started behind. Grok 1 (March 2024) had 8k tokens of context at a time when GPT-4 Turbo was already at 128k. Grok 3 expanded to 131k. Grok 4, released July 2025, jumped to 256k. And then in November 2025, Grok 4.1 reached 2M tokens, which put it above what OpenAI and Anthropic offered at the time (Grok 4.1 has since been retired). The April 2026 model, Grok 4.3, standardized on a 1M window at much lower pricing ($1.25 per million input).
That trajectory from 8k to 1M is one of the steepest context expansions any major AI lab has shipped. It matters because context length is one of the few hard capability limits that is genuinely visible to end users. An engineer deciding whether to use Grok vs Claude vs Gemini for analyzing a large codebase cares about whether the whole thing fits in one prompt more than about which model has a slightly higher benchmark score.
What 1 million tokens actually holds
Grok 4.3's 1M token window (about 750,000 words) is large enough to ingest things most AI models can't touch:
- The first four Harry Potter books (~460K words) with room left over
- War and Peace (~587K words) with room left over
- The Lord of the Rings plus The Hobbit (~576K words)
- A medium-sized application's codebase (on the order of 100K lines of code)
- Roughly 3,000 single-spaced pages at 250 words per page
Whether all 1M tokens are genuinely usable for complex reasoning is a separate question. Independent evaluations of long-context models tend to show some degradation past 500K tokens on synthesis tasks (where the model has to reason across the entire context), though retrieval-style tasks (finding specific facts) hold up better. This pattern matches what research groups have found with every long-context model, not just Grok.
Pricing
Grok's pricing tiers at September 2026 list rates:
| Model | Input / 1M tokens | Output / 1M tokens |
|---|---|---|
| Grok 4.7 | $2.00 | $6.00 |
| Grok 4.6 | $2.00 | $6.00 |
| Grok 4.5 | $2.00 | $6.00 |
| Grok 4.3 | $1.25 | $2.50 |
Grok 4.3 is the cheapest current model at $1.25 per million input tokens and also has the largest window (1M). Grok 4.7, 4.6 and 4.5 cost more ($2.00 input, $6.00 output) for a 500K window.
Consumer Grok on X vs the API
Grok is available three ways: inside the X (Twitter) app, through the Grok website, and via the xAI API. Limits differ across all three.
- X app (free): Grok access with strict daily message caps. Context appears capped well below the API maximum.
- SuperGrok ($30/month): Expanded Grok access, longer context per message, no hard caps.
- SuperGrok Heavy ($300/month): 16-agent mode, priority access, enterprise-style availability.
- API: Full specs listed in the tables above. Pay per token. Suitable for production workloads.
If you're testing whether your 100k-word document fits, the X app will reject you long before the API would. Use the API or a tool like OpenRouter for real benchmarks.
Check your prompt against every model
See how your text fits in Grok, ChatGPT, Claude, Gemini, and more, side by side
AI Prompt Word CounterFAQ
What is Grok's word limit?
Grok 4.7, the current flagship, accepts about 370,000 words (500K tokens). Grok 4.3 accepts about 750,000 words (1M tokens).
Is Grok's 1M token window real or marketing?
It is real for retrieval tasks, where Grok holds up well deep into the window. For complex multi-step reasoning across the full 1M, some degradation past 500K tokens is expected, matching the pattern seen with every long-context model.
Can Grok read entire book series?
Individual long books and trilogies fit on Grok 4.3 (1M tokens): Lord of the Rings plus The Hobbit fits, and War and Peace fits with room to spare. The full seven-book Harry Potter series (~1.08M words) is too large for any current Grok model.
How much does it cost to fill Grok 4.3's context?
At $1.25 per million input tokens, filling Grok 4.3's 1M window costs about $1.25 in input. Filling Grok 4.7's 500K window at $2 per million costs about $1.00.
How does Grok compare to Claude and GPT on context?
Grok 4.3 (1M) is in the same range as the 1M windows on Claude's top models and the roughly 1.05M windows in OpenAI's GPT-6 family, while Grok 4.7 and 4.6 take 500K. Quality within that context depends on the task and is disputed across benchmarks.